I guess you could do such a test on a turbo and collect the power and wheel speed data. Would make an interesting comparison. I’d probably go for a longer test too given that it may be helping to reduce fatigue. Think it would be interesting to test at different cadences and powers too, given that by spending less time in the “dead spot” it could be beneficial in high resistance and low cadence scenario.
I’ve got a spider (P2M) and a hub (PowerTap) based PM. PowerTap, being hub based, is meant to be accurate with oval or round rings. P2M, being spider based, is meant to be out. If I had the time I’d do a few tests back to back with both PM fitted and compare the difference.
My experience, I quite often switch between bikes with oval and non-oval. I don’t notice any difference when pedalling (though am convinced they are less fatiguing over the course of a ride.)
… having said that, one thing I do notice with the rotor rings is a change in orientation of the ring mounting. Comes with 5 pre-defined positions. If I go from, say, position 2 to position 3, the difference is noticeable.